We aim to solve the problem of how the geometry of cleavage is controlled in early frog and fish embryos. This is a classic, unsolved problem in cell biology. The lessons we learn from solving it will likely inform on how large animal cells are physically organized in general. The position of cleavage planes depends on the prior position of mitotic spindles, which in turn depends on growth and movement of large microtubule asters (radial arrays of microtubules initiated from centrosomes that grow to fill the cytoplasm). Using improved microscopy methods (immunofluorescence in frog and live imaging in fish) we have visualized novel aspects of aster morphology and movement in early embryos. These include formation of a zone of decreased microtubule density where neighboring asters meet each other in telophase, and aster expansion with morphology that is inconsistent with the standard model of plus end polymerization only. We propose to elucidate the molecular mechanisms that underlie these novel aspects of microtubule organization. Our main experimental system will be concentrated cytoplasmic extracts from frog eggs. These recapitulate key aspects of aster growth, morphology and interaction in a cell-free system that is excellent for imaging and molecular perturbation. Important conclusions will be tested in frog and fish embryos. In aim 1 we will determine the molecular mechanisms that prevent inter-penetration of microtubules from neighboring asters that are expanding in telophase, and generate a zone of reduced microtubule density between sister asters. In aim 2 we will determine how asters expand, quantifying the contribution of microtubule elongation, nucleation and sliding. In aim 3 we will determine how microtubules are nucleated at sites far from centrosomes, but near pre-existing microtubules, which is key to morphogenesis of the microtubule cytoskeleton in both interphase and mitosis in large egg cells. Based on preliminary imaging and perturbation data, we propose an original model for how centrosomes and mitotic spindles are positioned in early embryos. Key aspects of this model are (i) pulling force is exerted along the length of astral microtubules by dynein anchored in the cytoplasm, and (ii) microtubule length is limited by interaction of growing asters with the neighboring aster at the middle of the cell, and the cortex at the outside of the cell. We propose to test and extend this model. If confirmed, it will provide significant conceptual progress on early embryo organization.